linux/arch/x86/boot/compressed/head_64.S

567 lines
12 KiB
ArmAsm
Raw Normal View History

/*
* linux/boot/head.S
*
* Copyright (C) 1991, 1992, 1993 Linus Torvalds
*/
/*
* head.S contains the 32-bit startup code.
*
* NOTE!!! Startup happens at absolute address 0x00001000, which is also where
* the page directory will exist. The startup code will be overwritten by
* the page directory. [According to comments etc elsewhere on a compressed
* kernel it will end up at 0x1000 + 1Mb I hope so as I assume this. - AC]
*
* Page 0 is deliberately kept safe, since System Management Mode code in
* laptops may need to access the BIOS data stored there. This is also
* useful for future device drivers that either access the BIOS via VM86
* mode.
*/
/*
* High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996
*/
.code32
.text
#include <linux/init.h>
#include <linux/linkage.h>
#include <asm/segment.h>
#include <asm/boot.h>
#include <asm/msr.h>
#include <asm/processor-flags.h>
#include <asm/asm-offsets.h>
#include <asm/bootparam.h>
x86/build: Build compressed x86 kernels as PIE The 32-bit x86 assembler in binutils 2.26 will generate R_386_GOT32X relocation to get the symbol address in PIC. When the compressed x86 kernel isn't built as PIC, the linker optimizes R_386_GOT32X relocations to their fixed symbol addresses. However, when the compressed x86 kernel is loaded at a different address, it leads to the following load failure: Failed to allocate space for phdrs during the decompression stage. If the compressed x86 kernel is relocatable at run-time, it should be compiled with -fPIE, instead of -fPIC, if possible and should be built as Position Independent Executable (PIE) so that linker won't optimize R_386_GOT32X relocation to its fixed symbol address. Older linkers generate R_386_32 relocations against locally defined symbols, _bss, _ebss, _got and _egot, in PIE. It isn't wrong, just less optimal than R_386_RELATIVE. But the x86 kernel fails to properly handle R_386_32 relocations when relocating the kernel. To generate R_386_RELATIVE relocations, we mark _bss, _ebss, _got and _egot as hidden in both 32-bit and 64-bit x86 kernels. To build a 64-bit compressed x86 kernel as PIE, we need to disable the relocation overflow check to avoid relocation overflow errors. We do this with a new linker command-line option, -z noreloc-overflow, which got added recently: commit 4c10bbaa0912742322f10d9d5bb630ba4e15dfa7 Author: H.J. Lu <hjl.tools@gmail.com> Date: Tue Mar 15 11:07:06 2016 -0700 Add -z noreloc-overflow option to x86-64 ld Add -z noreloc-overflow command-line option to the x86-64 ELF linker to disable relocation overflow check. This can be used to avoid relocation overflow check if there will be no dynamic relocation overflow at run-time. The 64-bit compressed x86 kernel is built as PIE only if the linker supports -z noreloc-overflow. So far 64-bit relocatable compressed x86 kernel boots fine even when it is built as a normal executable. Signed-off-by: H.J. Lu <hjl.tools@gmail.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org [ Edited the changelog and comments. ] Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-03-17 03:04:35 +00:00
/*
* Locally defined symbols should be marked hidden:
*/
.hidden _bss
.hidden _ebss
.hidden _got
.hidden _egot
__HEAD
.code32
ENTRY(startup_32)
/*
* 32bit entry is 0 and it is ABI so immutable!
* If we come here directly from a bootloader,
* kernel(text+data+bss+brk) ramdisk, zero_page, command line
* all need to be under the 4G limit.
*/
cld
/*
* Test KEEP_SEGMENTS flag to see if the bootloader is asking
* us to not reload segments
*/
testb $KEEP_SEGMENTS, BP_loadflags(%esi)
jnz 1f
cli
movl $(__BOOT_DS), %eax
movl %eax, %ds
movl %eax, %es
movl %eax, %ss
1:
/*
* Calculate the delta between where we were compiled to run
* at and where we were actually loaded at. This can only be done
* with a short local call on x86. Nothing else will tell us what
* address we are running at. The reserved chunk of the real-mode
* data at 0x1e4 (defined as a scratch field) are used as the stack
* for this calculation. Only 4 bytes are needed.
*/
leal (BP_scratch+4)(%esi), %esp
call 1f
1: popl %ebp
subl $1b, %ebp
/* setup a stack and make sure cpu supports long mode. */
movl $boot_stack_end, %eax
addl %ebp, %eax
movl %eax, %esp
call verify_cpu
testl %eax, %eax
jnz no_longmode
/*
* Compute the delta between where we were compiled to run at
* and where the code will actually run at.
*
* %ebp contains the address we are loaded at by the boot loader and %ebx
* contains the address where we should move the kernel image temporarily
* for safe in-place decompression.
*/
#ifdef CONFIG_RELOCATABLE
movl %ebp, %ebx
movl BP_kernel_alignment(%esi), %eax
decl %eax
addl %eax, %ebx
notl %eax
andl %eax, %ebx
cmpl $LOAD_PHYSICAL_ADDR, %ebx
jge 1f
#endif
movl $LOAD_PHYSICAL_ADDR, %ebx
1:
/* Target address to relocate to for decompression */
x86/boot: Move compressed kernel to the end of the decompression buffer This change makes later calculations about where the kernel is located easier to reason about. To better understand this change, we must first clarify what 'VO' and 'ZO' are. These values were introduced in commits by hpa: 77d1a4999502 ("x86, boot: make symbols from the main vmlinux available") 37ba7ab5e33c ("x86, boot: make kernel_alignment adjustable; new bzImage fields") Specifically: All names prefixed with 'VO_': - relate to the uncompressed kernel image - the size of the VO image is: VO__end-VO__text ("VO_INIT_SIZE" define) All names prefixed with 'ZO_': - relate to the bootable compressed kernel image (boot/compressed/vmlinux), which is composed of the following memory areas: - head text - compressed kernel (VO image and relocs table) - decompressor code - the size of the ZO image is: ZO__end - ZO_startup_32 ("ZO_INIT_SIZE" define, though see below) The 'INIT_SIZE' value is used to find the larger of the two image sizes: #define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) #define VO_INIT_SIZE (VO__end - VO__text) #if ZO_INIT_SIZE > VO_INIT_SIZE # define INIT_SIZE ZO_INIT_SIZE #else # define INIT_SIZE VO_INIT_SIZE #endif The current code uses extract_offset to decide where to position the copied ZO (i.e. ZO starts at extract_offset). (This is why ZO_INIT_SIZE currently includes the extract_offset.) Why does z_extract_offset exist? It's needed because we are trying to minimize the amount of RAM used for the whole act of creating an uncompressed, executable, properly relocation-linked kernel image in system memory. We do this so that kernels can be booted on even very small systems. To achieve the goal of minimal memory consumption we have implemented an in-place decompression strategy: instead of cleanly separating the VO and ZO images and also allocating some memory for the decompression code's runtime needs, we instead create this elaborate layout of memory buffers where the output (decompressed) stream, as it progresses, overlaps with and destroys the input (compressed) stream. This can only be done safely if the ZO image is placed to the end of the VO range, plus a certain amount of safety distance to make sure that when the last bytes of the VO range are decompressed, the compressed stream pointer is safely beyond the end of the VO range. z_extract_offset is calculated in arch/x86/boot/compressed/mkpiggy.c during the build process, at a point when we know the exact compressed and uncompressed size of the kernel images and can calculate this safe minimum offset value. (Note that the mkpiggy.c calculation is not perfect, because we don't know the decompressor used at that stage, so the z_extract_offset calculation is necessarily imprecise and is mostly based on gzip internals - we'll improve that in the next patch.) When INIT_SIZE is bigger than VO_INIT_SIZE (uncommon but possible), the copied ZO occupies the memory from extract_offset to the end of decompression buffer. It overlaps with the soon-to-be-uncompressed kernel like this: |-----compressed kernel image------| V V 0 extract_offset +INIT_SIZE |-----------|---------------|-------------------------|--------| | | | | VO__text startup_32 of ZO VO__end ZO__end ^ ^ |-------uncompressed kernel image---------| When INIT_SIZE is equal to VO_INIT_SIZE (likely) there's still space left from end of ZO to the end of decompressing buffer, like below. |-compressed kernel image-| V V 0 extract_offset +INIT_SIZE |-----------|---------------|-------------------------|--------| | | | | VO__text startup_32 of ZO ZO__end VO__end ^ ^ |------------uncompressed kernel image-------------| To simplify calculations and avoid special cases, it is cleaner to always place the compressed kernel image in memory so that ZO__end is at the end of the decompression buffer, instead of placing t at the start of extract_offset as is currently done. This patch adds BP_init_size (which is the INIT_SIZE as passed in from the boot_params) into asm-offsets.c to make it visible to the assembly code. Then when moving the ZO, it calculates the starting position of the copied ZO (via BP_init_size and the ZO run size) so that the VO__end will be at the end of the decompression buffer. To make the position calculation safe, the end of ZO is page aligned (and a comment is added to the existing VO alignment for good measure). Signed-off-by: Yinghai Lu <yinghai@kernel.org> [ Rewrote changelog and comments. ] Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Baoquan He <bhe@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Young <dyoung@redhat.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: lasse.collin@tukaani.org Link: http://lkml.kernel.org/r/1461888548-32439-3-git-send-email-keescook@chromium.org [ Rewrote the changelog some more. ] Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-04-29 00:09:04 +00:00
movl BP_init_size(%esi), %eax
subl $_end, %eax
addl %eax, %ebx
/*
* Prepare for entering 64 bit mode
*/
/* Load new GDT with the 64bit segments using 32bit descriptor */
addl %ebp, gdt+2(%ebp)
lgdt gdt(%ebp)
/* Enable PAE mode */
movl %cr4, %eax
orl $X86_CR4_PAE, %eax
movl %eax, %cr4
/*
* Build early 4G boot pagetable
*/
/* Initialize Page tables to 0 */
leal pgtable(%ebx), %edi
xorl %eax, %eax
x86/KASLR: Build identity mappings on demand Currently KASLR only supports relocation in a small physical range (from 16M to 1G), due to using the initial kernel page table identity mapping. To support ranges above this, we need to have an identity mapping for the desired memory range before we can decompress (and later run) the kernel. 32-bit kernels already have the needed identity mapping. This patch adds identity mappings for the needed memory ranges on 64-bit kernels. This happens in two possible boot paths: If loaded via startup_32(), we need to set up the needed identity map. If loaded from a 64-bit bootloader, the bootloader will have already set up an identity mapping, and we'll start via the compressed kernel's startup_64(). In this case, the bootloader's page tables need to be avoided while selecting the new uncompressed kernel location. If not, the decompressor could overwrite them during decompression. To accomplish this, we could walk the pagetable and find every page that is used, and add them to mem_avoid, but this needs extra code and will require increasing the size of the mem_avoid array. Instead, we can create a new set of page tables for our own identity mapping instead. The pages for the new page table will come from the _pagetable section of the compressed kernel, which means they are already contained by in mem_avoid array. To do this, we reuse the code from the uncompressed kernel's identity mapping routines. The _pgtable will be shared by both the 32-bit and 64-bit paths to reduce init_size, as now the compressed kernel's _rodata to _end will contribute to init_size. To handle the possible mappings, we need to increase the existing page table buffer size: When booting via startup_64(), we need to cover the old VO, params, cmdline and uncompressed kernel. In an extreme case we could have them all beyond the 512G boundary, which needs (2+2)*4 pages with 2M mappings. And we'll need 2 for first 2M for VGA RAM. One more is needed for level4. This gets us to 19 pages total. When booting via startup_32(), KASLR could move the uncompressed kernel above 4G, so we need to create extra identity mappings, which should only need (2+2) pages at most when it is beyond the 512G boundary. So 19 pages is sufficient for this case as well. The resulting BOOT_*PGT_SIZE defines use the "_SIZE" suffix on their names to maintain logical consistency with the existing BOOT_HEAP_SIZE and BOOT_STACK_SIZE defines. This patch is based on earlier patches from Yinghai Lu and Baoquan He. Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Baoquan He <bhe@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Borislav Petkov <bp@suse.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Young <dyoung@redhat.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: Yinghai Lu <yinghai@kernel.org> Cc: kernel-hardening@lists.openwall.com Cc: lasse.collin@tukaani.org Link: http://lkml.kernel.org/r/1462572095-11754-4-git-send-email-keescook@chromium.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-06 22:01:35 +00:00
movl $(BOOT_INIT_PGT_SIZE/4), %ecx
rep stosl
/* Build Level 4 */
leal pgtable + 0(%ebx), %edi
leal 0x1007 (%edi), %eax
movl %eax, 0(%edi)
/* Build Level 3 */
leal pgtable + 0x1000(%ebx), %edi
leal 0x1007(%edi), %eax
movl $4, %ecx
1: movl %eax, 0x00(%edi)
addl $0x00001000, %eax
addl $8, %edi
decl %ecx
jnz 1b
/* Build Level 2 */
leal pgtable + 0x2000(%ebx), %edi
movl $0x00000183, %eax
movl $2048, %ecx
1: movl %eax, 0(%edi)
addl $0x00200000, %eax
addl $8, %edi
decl %ecx
jnz 1b
/* Enable the boot page tables */
leal pgtable(%ebx), %eax
movl %eax, %cr3
/* Enable Long mode in EFER (Extended Feature Enable Register) */
movl $MSR_EFER, %ecx
rdmsr
btsl $_EFER_LME, %eax
wrmsr
/* After gdt is loaded */
xorl %eax, %eax
lldt %ax
movl $__BOOT_TSS, %eax
ltr %ax
/*
* Setup for the jump to 64bit mode
*
* When the jump is performend we will be in long mode but
* in 32bit compatibility mode with EFER.LME = 1, CS.L = 0, CS.D = 1
* (and in turn EFER.LMA = 1). To jump into 64bit mode we use
* the new gdt/idt that has __KERNEL_CS with CS.L = 1.
* We place all of the values on our mini stack so lret can
* used to perform that far jump.
*/
pushl $__KERNEL_CS
leal startup_64(%ebp), %eax
2014-01-10 15:54:31 +00:00
#ifdef CONFIG_EFI_MIXED
movl efi32_config(%ebp), %ebx
cmp $0, %ebx
jz 1f
leal handover_entry(%ebp), %eax
1:
#endif
pushl %eax
/* Enter paged protected Mode, activating Long Mode */
movl $(X86_CR0_PG | X86_CR0_PE), %eax /* Enable Paging and Protected mode */
movl %eax, %cr0
/* Jump from 32bit compatibility mode into 64bit mode. */
lret
ENDPROC(startup_32)
2014-01-10 15:54:31 +00:00
#ifdef CONFIG_EFI_MIXED
.org 0x190
ENTRY(efi32_stub_entry)
add $0x4, %esp /* Discard return address */
popl %ecx
popl %edx
popl %esi
leal (BP_scratch+4)(%esi), %esp
call 1f
1: pop %ebp
subl $1b, %ebp
movl %ecx, efi32_config(%ebp)
movl %edx, efi32_config+8(%ebp)
sgdtl efi32_boot_gdt(%ebp)
leal efi32_config(%ebp), %eax
movl %eax, efi_config(%ebp)
jmp startup_32
ENDPROC(efi32_stub_entry)
#endif
.code64
.org 0x200
ENTRY(startup_64)
/*
* 64bit entry is 0x200 and it is ABI so immutable!
* We come here either from startup_32 or directly from a
* 64bit bootloader.
* If we come here from a bootloader, kernel(text+data+bss+brk),
* ramdisk, zero_page, command line could be above 4G.
* We depend on an identity mapped page table being provided
* that maps our entire kernel(text+data+bss+brk), zero page
* and command line.
*/
/* Setup data segments. */
xorl %eax, %eax
movl %eax, %ds
movl %eax, %es
movl %eax, %ss
movl %eax, %fs
movl %eax, %gs
/*
* Compute the decompressed kernel start address. It is where
* we were loaded at aligned to a 2M boundary. %rbp contains the
* decompressed kernel start address.
*
* If it is a relocatable kernel then decompress and run the kernel
* from load address aligned to 2MB addr, otherwise decompress and
* run the kernel from LOAD_PHYSICAL_ADDR
*
* We cannot rely on the calculation done in 32-bit mode, since we
* may have been invoked via the 64-bit entry point.
*/
/* Start with the delta to where the kernel will run at. */
#ifdef CONFIG_RELOCATABLE
leaq startup_32(%rip) /* - $startup_32 */, %rbp
movl BP_kernel_alignment(%rsi), %eax
decl %eax
addq %rax, %rbp
notq %rax
andq %rax, %rbp
cmpq $LOAD_PHYSICAL_ADDR, %rbp
jge 1f
#endif
movq $LOAD_PHYSICAL_ADDR, %rbp
1:
/* Target address to relocate to for decompression */
x86/boot: Move compressed kernel to the end of the decompression buffer This change makes later calculations about where the kernel is located easier to reason about. To better understand this change, we must first clarify what 'VO' and 'ZO' are. These values were introduced in commits by hpa: 77d1a4999502 ("x86, boot: make symbols from the main vmlinux available") 37ba7ab5e33c ("x86, boot: make kernel_alignment adjustable; new bzImage fields") Specifically: All names prefixed with 'VO_': - relate to the uncompressed kernel image - the size of the VO image is: VO__end-VO__text ("VO_INIT_SIZE" define) All names prefixed with 'ZO_': - relate to the bootable compressed kernel image (boot/compressed/vmlinux), which is composed of the following memory areas: - head text - compressed kernel (VO image and relocs table) - decompressor code - the size of the ZO image is: ZO__end - ZO_startup_32 ("ZO_INIT_SIZE" define, though see below) The 'INIT_SIZE' value is used to find the larger of the two image sizes: #define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) #define VO_INIT_SIZE (VO__end - VO__text) #if ZO_INIT_SIZE > VO_INIT_SIZE # define INIT_SIZE ZO_INIT_SIZE #else # define INIT_SIZE VO_INIT_SIZE #endif The current code uses extract_offset to decide where to position the copied ZO (i.e. ZO starts at extract_offset). (This is why ZO_INIT_SIZE currently includes the extract_offset.) Why does z_extract_offset exist? It's needed because we are trying to minimize the amount of RAM used for the whole act of creating an uncompressed, executable, properly relocation-linked kernel image in system memory. We do this so that kernels can be booted on even very small systems. To achieve the goal of minimal memory consumption we have implemented an in-place decompression strategy: instead of cleanly separating the VO and ZO images and also allocating some memory for the decompression code's runtime needs, we instead create this elaborate layout of memory buffers where the output (decompressed) stream, as it progresses, overlaps with and destroys the input (compressed) stream. This can only be done safely if the ZO image is placed to the end of the VO range, plus a certain amount of safety distance to make sure that when the last bytes of the VO range are decompressed, the compressed stream pointer is safely beyond the end of the VO range. z_extract_offset is calculated in arch/x86/boot/compressed/mkpiggy.c during the build process, at a point when we know the exact compressed and uncompressed size of the kernel images and can calculate this safe minimum offset value. (Note that the mkpiggy.c calculation is not perfect, because we don't know the decompressor used at that stage, so the z_extract_offset calculation is necessarily imprecise and is mostly based on gzip internals - we'll improve that in the next patch.) When INIT_SIZE is bigger than VO_INIT_SIZE (uncommon but possible), the copied ZO occupies the memory from extract_offset to the end of decompression buffer. It overlaps with the soon-to-be-uncompressed kernel like this: |-----compressed kernel image------| V V 0 extract_offset +INIT_SIZE |-----------|---------------|-------------------------|--------| | | | | VO__text startup_32 of ZO VO__end ZO__end ^ ^ |-------uncompressed kernel image---------| When INIT_SIZE is equal to VO_INIT_SIZE (likely) there's still space left from end of ZO to the end of decompressing buffer, like below. |-compressed kernel image-| V V 0 extract_offset +INIT_SIZE |-----------|---------------|-------------------------|--------| | | | | VO__text startup_32 of ZO ZO__end VO__end ^ ^ |------------uncompressed kernel image-------------| To simplify calculations and avoid special cases, it is cleaner to always place the compressed kernel image in memory so that ZO__end is at the end of the decompression buffer, instead of placing t at the start of extract_offset as is currently done. This patch adds BP_init_size (which is the INIT_SIZE as passed in from the boot_params) into asm-offsets.c to make it visible to the assembly code. Then when moving the ZO, it calculates the starting position of the copied ZO (via BP_init_size and the ZO run size) so that the VO__end will be at the end of the decompression buffer. To make the position calculation safe, the end of ZO is page aligned (and a comment is added to the existing VO alignment for good measure). Signed-off-by: Yinghai Lu <yinghai@kernel.org> [ Rewrote changelog and comments. ] Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Baoquan He <bhe@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Young <dyoung@redhat.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: lasse.collin@tukaani.org Link: http://lkml.kernel.org/r/1461888548-32439-3-git-send-email-keescook@chromium.org [ Rewrote the changelog some more. ] Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-04-29 00:09:04 +00:00
movl BP_init_size(%rsi), %ebx
subl $_end, %ebx
addq %rbp, %rbx
/* Set up the stack */
leaq boot_stack_end(%rbx), %rsp
x86/boot/compressed: Enable 5-level paging during decompression stage We need to cover two basic cases: when bootloader left us in 32-bit mode and when bootloader enabled long mode. The patch implements unified codepath to enabled 5-level paging for both cases. It means case when we start in 32-bit mode, we first enable long mode with 4-level and then switch over to 5-level paging. Switching from 4-level to 5-level paging is not trivial. We cannot do it directly. Setting LA57 in long mode would trigger #GP. So we need to switch off long mode first and the then re-enable with 5-level paging. NOTE: The need of switching off long mode means we are in trouble if bootloader put us above 4G boundary. If bootloader wants to boot 5-level paging kernel, it has to put kernel below 4G or enable 5-level paging on it's own, so we could avoid the step. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20170606113133.22974-7-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-06-06 11:31:25 +00:00
#ifdef CONFIG_X86_5LEVEL
/* Check if 5-level paging has already enabled */
movq %cr4, %rax
testl $X86_CR4_LA57, %eax
jnz lvl5
/*
* At this point we are in long mode with 4-level paging enabled,
* but we want to enable 5-level paging.
*
* The problem is that we cannot do it directly. Setting LA57 in
* long mode would trigger #GP. So we need to switch off long mode
* first.
*
* NOTE: This is not going to work if bootloader put us above 4G
* limit.
*
* The first step is go into compatibility mode.
*/
/* Clear additional page table */
leaq lvl5_pgtable(%rbx), %rdi
xorq %rax, %rax
movq $(PAGE_SIZE/8), %rcx
rep stosq
/*
* Setup current CR3 as the first and only entry in a new top level
* page table.
*/
movq %cr3, %rdi
leaq 0x7 (%rdi), %rax
movq %rax, lvl5_pgtable(%rbx)
/* Switch to compatibility mode (CS.L = 0 CS.D = 1) via far return */
pushq $__KERNEL32_CS
leaq compatible_mode(%rip), %rax
pushq %rax
lretq
lvl5:
#endif
/* Zero EFLAGS */
pushq $0
popfq
/*
* Copy the compressed kernel to the end of our buffer
* where decompression in place becomes safe.
*/
pushq %rsi
leaq (_bss-8)(%rip), %rsi
leaq (_bss-8)(%rbx), %rdi
movq $_bss /* - $startup_32 */, %rcx
shrq $3, %rcx
std
rep movsq
cld
popq %rsi
/*
* Jump to the relocated address.
*/
leaq relocated(%rbx), %rax
jmp *%rax
2014-01-10 15:54:31 +00:00
#ifdef CONFIG_EFI_STUB
x86/boot/64: Extract efi_pe_entry() from startup_64() Similarly to the 32-bit code, efi_pe_entry body() is somehow squashed into startup_64(). In the old days, we forced startup_64() to start at offset 0x200 and efi_pe_entry() to start at 0x210. But this requirement was removed long time ago, in: 99f857db8857 ("x86, build: Dynamically find entry points in compressed startup code") The way it is now makes the code less readable and illogical. Given we can now safely extract the inlined efi_pe_entry() body from startup_64() into a separate function, we do so. We also annotate the function appropriatelly by ENTRY+ENDPROC. ABI offsets are preserved: 0000000000000000 T startup_32 0000000000000200 T startup_64 0000000000000390 T efi64_stub_entry On the top-level, it looked like: .org 0x200 ENTRY(startup_64) #ifdef CONFIG_EFI_STUB ; start of inlined jmp preferred_addr GLOBAL(efi_pe_entry) ... ; a lot of assembly (efi_pe_entry) leaq preferred_addr(%rax), %rax jmp *%rax preferred_addr: #endif ; end of inlined ... ; a lot of assembly (startup_64) ENDPROC(startup_64) And it is now converted into: .org 0x200 ENTRY(startup_64) ... ; a lot of assembly (startup_64) ENDPROC(startup_64) #ifdef CONFIG_EFI_STUB ENTRY(efi_pe_entry) ... ; a lot of assembly (efi_pe_entry) leaq startup_64(%rax), %rax jmp *%rax ENDPROC(efi_pe_entry) #endif Signed-off-by: Jiri Slaby <jslaby@suse.cz> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Matt Fleming <matt@codeblueprint.co.uk> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: ard.biesheuvel@linaro.org Cc: linux-efi@vger.kernel.org Link: http://lkml.kernel.org/r/20170824073327.4129-2-jslaby@suse.cz Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-24 07:33:27 +00:00
/* The entry point for the PE/COFF executable is efi_pe_entry. */
ENTRY(efi_pe_entry)
movq %rcx, efi64_config(%rip) /* Handle */
movq %rdx, efi64_config+8(%rip) /* EFI System table pointer */
leaq efi64_config(%rip), %rax
movq %rax, efi_config(%rip)
call 1f
1: popq %rbp
subq $1b, %rbp
/*
* Relocate efi_config->call().
*/
addq %rbp, efi64_config+40(%rip)
movq %rax, %rdi
call make_boot_params
cmpq $0,%rax
je fail
mov %rax, %rsi
leaq startup_32(%rip), %rax
movl %eax, BP_code32_start(%rsi)
jmp 2f /* Skip the relocation */
handover_entry:
call 1f
1: popq %rbp
subq $1b, %rbp
/*
* Relocate efi_config->call().
*/
movq efi_config(%rip), %rax
addq %rbp, 40(%rax)
2:
movq efi_config(%rip), %rdi
call efi_main
movq %rax,%rsi
cmpq $0,%rax
jne 2f
fail:
/* EFI init failed, so hang. */
hlt
jmp fail
2:
movl BP_code32_start(%esi), %eax
leaq startup_64(%rax), %rax
jmp *%rax
ENDPROC(efi_pe_entry)
2014-01-10 15:54:31 +00:00
.org 0x390
ENTRY(efi64_stub_entry)
movq %rdi, efi64_config(%rip) /* Handle */
movq %rsi, efi64_config+8(%rip) /* EFI System table pointer */
leaq efi64_config(%rip), %rax
movq %rax, efi_config(%rip)
movq %rdx, %rsi
jmp handover_entry
ENDPROC(efi64_stub_entry)
#endif
.text
relocated:
/*
* Clear BSS (stack is currently empty)
*/
xorl %eax, %eax
leaq _bss(%rip), %rdi
leaq _ebss(%rip), %rcx
subq %rdi, %rcx
shrq $3, %rcx
rep stosq
/*
* Adjust our own GOT
*/
leaq _got(%rip), %rdx
leaq _egot(%rip), %rcx
1:
cmpq %rcx, %rdx
jae 2f
addq %rbx, (%rdx)
addq $8, %rdx
jmp 1b
2:
/*
* Do the extraction, and jump to the new kernel..
*/
pushq %rsi /* Save the real mode argument */
movq %rsi, %rdi /* real mode address */
leaq boot_heap(%rip), %rsi /* malloc area for uncompression */
leaq input_data(%rip), %rdx /* input_data */
movl $z_input_len, %ecx /* input_len */
movq %rbp, %r8 /* output target address */
x86, kaslr: Prevent .bss from overlaping initrd When choosing a random address, the current implementation does not take into account the reversed space for .bss and .brk sections. Thus the relocated kernel may overlap other components in memory. Here is an example of the overlap from a x86_64 kernel in qemu (the ranges of physical addresses are presented): Physical Address 0x0fe00000 --+--------------------+ <-- randomized base / | relocated kernel | vmlinux.bin | (from vmlinux.bin) | 0x1336d000 (an ELF file) +--------------------+-- \ | | \ 0x1376d870 --+--------------------+ | | relocs table | | 0x13c1c2a8 +--------------------+ .bss and .brk | | | 0x13ce6000 +--------------------+ | | | / 0x13f77000 | initrd |-- | | 0x13fef374 +--------------------+ The initrd image will then be overwritten by the memset during early initialization: [ 1.655204] Unpacking initramfs... [ 1.662831] Initramfs unpacking failed: junk in compressed archive This patch prevents the above situation by requiring a larger space when looking for a random kernel base, so that existing logic can effectively avoids the overlap. [kees: switched to perl to avoid hex translation pain in mawk vs gawk] [kees: calculated overlap without relocs table] Fixes: 82fa9637a2 ("x86, kaslr: Select random position from e820 maps") Reported-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Junjie Mao <eternal.n08@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Josh Triplett <josh@joshtriplett.org> Cc: Matt Fleming <matt.fleming@intel.com> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1414762838-13067-1-git-send-email-eternal.n08@gmail.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2014-10-31 13:40:38 +00:00
movq $z_output_len, %r9 /* decompressed length, end of relocs */
call extract_kernel /* returns kernel location in %rax */
popq %rsi
/*
* Jump to the decompressed kernel.
*/
jmp *%rax
.code32
x86/boot/compressed: Enable 5-level paging during decompression stage We need to cover two basic cases: when bootloader left us in 32-bit mode and when bootloader enabled long mode. The patch implements unified codepath to enabled 5-level paging for both cases. It means case when we start in 32-bit mode, we first enable long mode with 4-level and then switch over to 5-level paging. Switching from 4-level to 5-level paging is not trivial. We cannot do it directly. Setting LA57 in long mode would trigger #GP. So we need to switch off long mode first and the then re-enable with 5-level paging. NOTE: The need of switching off long mode means we are in trouble if bootloader put us above 4G boundary. If bootloader wants to boot 5-level paging kernel, it has to put kernel below 4G or enable 5-level paging on it's own, so we could avoid the step. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20170606113133.22974-7-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-06-06 11:31:25 +00:00
#ifdef CONFIG_X86_5LEVEL
compatible_mode:
/* Setup data and stack segments */
movl $__KERNEL_DS, %eax
movl %eax, %ds
movl %eax, %ss
/* Disable paging */
movl %cr0, %eax
btrl $X86_CR0_PG_BIT, %eax
movl %eax, %cr0
/* Point CR3 to 5-level paging */
leal lvl5_pgtable(%ebx), %eax
movl %eax, %cr3
/* Enable PAE and LA57 mode */
movl %cr4, %eax
orl $(X86_CR4_PAE | X86_CR4_LA57), %eax
movl %eax, %cr4
/* Calculate address we are running at */
call 1f
1: popl %edi
subl $1b, %edi
/* Prepare stack for far return to Long Mode */
pushl $__KERNEL_CS
leal lvl5(%edi), %eax
push %eax
/* Enable paging back */
movl $(X86_CR0_PG | X86_CR0_PE), %eax
movl %eax, %cr0
lret
#endif
no_longmode:
/* This isn't an x86-64 CPU so hang */
1:
hlt
jmp 1b
#include "../../kernel/verify_cpu.S"
.data
gdt:
.word gdt_end - gdt
.long gdt
.word 0
x86/boot/compressed: Enable 5-level paging during decompression stage We need to cover two basic cases: when bootloader left us in 32-bit mode and when bootloader enabled long mode. The patch implements unified codepath to enabled 5-level paging for both cases. It means case when we start in 32-bit mode, we first enable long mode with 4-level and then switch over to 5-level paging. Switching from 4-level to 5-level paging is not trivial. We cannot do it directly. Setting LA57 in long mode would trigger #GP. So we need to switch off long mode first and the then re-enable with 5-level paging. NOTE: The need of switching off long mode means we are in trouble if bootloader put us above 4G boundary. If bootloader wants to boot 5-level paging kernel, it has to put kernel below 4G or enable 5-level paging on it's own, so we could avoid the step. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20170606113133.22974-7-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-06-06 11:31:25 +00:00
.quad 0x00cf9a000000ffff /* __KERNEL32_CS */
.quad 0x00af9a000000ffff /* __KERNEL_CS */
.quad 0x00cf92000000ffff /* __KERNEL_DS */
.quad 0x0080890000000000 /* TS descriptor */
.quad 0x0000000000000000 /* TS continued */
gdt_end:
#ifdef CONFIG_EFI_STUB
efi_config:
.quad 0
2014-01-10 15:54:31 +00:00
#ifdef CONFIG_EFI_MIXED
.global efi32_config
efi32_config:
.fill 5,8,0
2014-01-10 15:54:31 +00:00
.quad efi64_thunk
.byte 0
#endif
.global efi64_config
efi64_config:
.fill 5,8,0
.quad efi_call
.byte 1
#endif /* CONFIG_EFI_STUB */
/*
* Stack and heap for uncompression
*/
.bss
.balign 4
boot_heap:
.fill BOOT_HEAP_SIZE, 1, 0
boot_stack:
.fill BOOT_STACK_SIZE, 1, 0
boot_stack_end:
/*
* Space for page tables (not in .bss so not zeroed)
*/
.section ".pgtable","a",@nobits
.balign 4096
pgtable:
x86/KASLR: Build identity mappings on demand Currently KASLR only supports relocation in a small physical range (from 16M to 1G), due to using the initial kernel page table identity mapping. To support ranges above this, we need to have an identity mapping for the desired memory range before we can decompress (and later run) the kernel. 32-bit kernels already have the needed identity mapping. This patch adds identity mappings for the needed memory ranges on 64-bit kernels. This happens in two possible boot paths: If loaded via startup_32(), we need to set up the needed identity map. If loaded from a 64-bit bootloader, the bootloader will have already set up an identity mapping, and we'll start via the compressed kernel's startup_64(). In this case, the bootloader's page tables need to be avoided while selecting the new uncompressed kernel location. If not, the decompressor could overwrite them during decompression. To accomplish this, we could walk the pagetable and find every page that is used, and add them to mem_avoid, but this needs extra code and will require increasing the size of the mem_avoid array. Instead, we can create a new set of page tables for our own identity mapping instead. The pages for the new page table will come from the _pagetable section of the compressed kernel, which means they are already contained by in mem_avoid array. To do this, we reuse the code from the uncompressed kernel's identity mapping routines. The _pgtable will be shared by both the 32-bit and 64-bit paths to reduce init_size, as now the compressed kernel's _rodata to _end will contribute to init_size. To handle the possible mappings, we need to increase the existing page table buffer size: When booting via startup_64(), we need to cover the old VO, params, cmdline and uncompressed kernel. In an extreme case we could have them all beyond the 512G boundary, which needs (2+2)*4 pages with 2M mappings. And we'll need 2 for first 2M for VGA RAM. One more is needed for level4. This gets us to 19 pages total. When booting via startup_32(), KASLR could move the uncompressed kernel above 4G, so we need to create extra identity mappings, which should only need (2+2) pages at most when it is beyond the 512G boundary. So 19 pages is sufficient for this case as well. The resulting BOOT_*PGT_SIZE defines use the "_SIZE" suffix on their names to maintain logical consistency with the existing BOOT_HEAP_SIZE and BOOT_STACK_SIZE defines. This patch is based on earlier patches from Yinghai Lu and Baoquan He. Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Baoquan He <bhe@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Borislav Petkov <bp@suse.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Young <dyoung@redhat.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: Yinghai Lu <yinghai@kernel.org> Cc: kernel-hardening@lists.openwall.com Cc: lasse.collin@tukaani.org Link: http://lkml.kernel.org/r/1462572095-11754-4-git-send-email-keescook@chromium.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-06 22:01:35 +00:00
.fill BOOT_PGT_SIZE, 1, 0
x86/boot/compressed: Enable 5-level paging during decompression stage We need to cover two basic cases: when bootloader left us in 32-bit mode and when bootloader enabled long mode. The patch implements unified codepath to enabled 5-level paging for both cases. It means case when we start in 32-bit mode, we first enable long mode with 4-level and then switch over to 5-level paging. Switching from 4-level to 5-level paging is not trivial. We cannot do it directly. Setting LA57 in long mode would trigger #GP. So we need to switch off long mode first and the then re-enable with 5-level paging. NOTE: The need of switching off long mode means we are in trouble if bootloader put us above 4G boundary. If bootloader wants to boot 5-level paging kernel, it has to put kernel below 4G or enable 5-level paging on it's own, so we could avoid the step. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20170606113133.22974-7-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-06-06 11:31:25 +00:00
#ifdef CONFIG_X86_5LEVEL
lvl5_pgtable:
.fill PAGE_SIZE, 1, 0
#endif